We are testing the hypothesis that families of proteins give rise to neuronal specifity. In the leech, monoclonal antibodies (mAbs) have shown that single types of neurons or sets of neurons are specified through individual molecular markers. Is this chemical labeling carried out by many unrelated molecules or is some of it carried out by protein families? Potential members of a 130 K protein family were identified by two mAbs that label different small sets of neurons. To test whether or not these proteins are part of a larger family, we will generate mAbs against 130 K proteins excised from SDS-Acrylamide gels. Other proteins specific for sets of neurons will be used as starting points in the search for families of proteins. We are developing the methodology to test whether a particular antigen belongs to a class of proteins. So far, we have three criteria: 1) antibodies must bind to antigens with a molecular weight characteristics of that class, 2) there must be mAbs that bind to all members of the family because they bind to a part of the proteins common to that family (allotypic mAbs), and 3) there must be mAbs that bind to just one protein because they bind to the part which is unique (idiotypic mAbs). In histological staining, mAbs that bind to the specific part of a protein will label subsets of neurons stained by mAbs that bind to the region of the proteins common to the family. When large, readily identifiable neurons are included in the staining pattern of mAbs pointing to a protein family, these neurons will be collected into pure pools and assayed for idiotypic and allotypic binding. The leech nervous system with its relatively small number of neurons is well suited for a comprehensive search into the molecular basis of neuronal specificity.